Water planning decisions are only as good asour ability to explain historical trends and make reasonablepredictions of future water availability. But predicting wa-ter availability can be a challenge in rapidly growing re-gions, where human modifications of land and waterscapesare changing the hydrologic system. Yet, many regions ofthe world lack the long-term hydrologic monitoring recordsneeded to understand past changes and predict future trends.We investigated this “predictions under change” problemin the data-scarce Thippagondanahalli (TG Halli) catchmentof the Arkavathy sub-basin in southern India. Inflows intoTG Halli reservoir have declined sharply since the 1970s.The causes of the drying are poorly understood, resulting inmisdirected or counter-productive management responses.Five plausible hypotheses that could explain the declinewere tested using data from field surveys and secondarysources: (1) changes in rainfall amount, seasonality and in-tensity; (2) increases in temperature; (3) groundwater ex-traction; (4) expansion of eucalyptus plantations; and (5)fragmentation of the river channel. Our results suggest thatgroundwater pumping, expansion of eucalyptus plantationsand, to a lesser extent, channel fragmentation are much morelikely to have caused the decline in surface flows in the TGHalli catchment than changing climate.The multiple-hypothesis approach presents a systematicway to quantify the relative contributions of proximate an-thropogenic and climate drivers to hydrological change. Theapproach not only makes a meaningful contribution to thepolicy debate but also helps prioritize and design future re-search. The approach is a first step to conducting use-inspiredsocio-hydrologic research in a watershed.